On the Implementation of a Reinforcement Learning-based Capacity Sharing Algorithm in O-RAN

TL;DR

This paper discusses implementing a reinforcement learning-based capacity sharing algorithm within the O-RAN architecture, including system operation, containerization, and validation through a testbed, addressing practical deployment challenges.

Contribution

It presents the first practical implementation of a RL-based capacity sharing algorithm in O-RAN, detailing system interfaces, containerization, and validation results.

Findings

The implementation demonstrates effective capacity sharing in O-RAN.

Performance validation shows promising results for the RL algorithm.

Insights into system operation and containerization are provided.

Abstract

The capacity sharing problem in Radio Access Network (RAN) slicing deals with the distribution of the capacity available in each RAN node among various RAN slices to satisfy their traffic demands and efficiently use the radio resources. While several capacity sharing algorithmic solutions have been proposed in the literature, their practical implementation still remains as a gap. In this paper, the implementation of a Reinforcement Learning-based capacity sharing algorithm over the O-RAN architecture is discussed, providing insights into the operation of the involved interfaces and the containerization of the solution. Moreover, the description of the testbed implemented to validate the solution is included and some performance and validation results are presented.